17 Surgical Management of Thoracolumbar Spinal Metastases Using Navigation

10.1055/b-0039-172728

17 Surgical Management of Thoracolumbar Spinal Metastases Using Navigation

Zach Pennington, A. Karim Ahmed, Camilo A. Molina, and Daniel M. Sciubba

Abstract

Given the significant frailty observed in many patients with spinal metastases, there has been increased emphasis on reducing surgical morbidity. One potential means of achieving this goal is by using minimally invasive surgery (MIS) techniques, such as mini-open approaches with percutaneous instrumentation, as they promise lower intraoperative blood loss and faster postoperative recovery. With advancements in intraoperative navigation, e.g., CT-guided navigation, more and more surgeons are able to add these techniques to their operative armamentarium. Here we describe the use of intraoperative navigation for the surgical treatment of spinal metastases and provide a comparison of open and MIS techniques for the treatment of this clinical pathology.

17.1 Introduction

Each year some 1.7 million Americans will be diagnosed with cancer ¹ and 40 to 70% of these patients ² ^, ³ ^, ⁴ ^, ⁵ ^, ⁶ ^, ⁷ will experience one or more spinal metastases. The majority of these lesions remain clinically silent; however, a small minority of patients—18,000 to 25,000 per year ⁸ ^, ⁹ ^, ¹⁰ ^, ¹¹ ^, ¹² ^, ¹³ ^, ¹⁴ —will present with some indication for surgery, whether it be neurological dysfunction ⁵ ^, ¹⁵ ^, ¹⁶ ^, ¹⁷ ^, ¹⁸ ^, ¹⁹ ^, ²⁰ ^, ²¹ ^, ²² ^, ²³ ^, ²⁴ ^, ²⁵ ^, ²⁶ or gross spinal instability. ⁸ ^, ¹⁵ ^, ²⁷ For these patients, surgical management has been demonstrated to provide superior treatment outcomes as compared to radiation or chemotherapy alone. ⁸ However, many patients within this population are particularly frail, and may not have the physical reserve to stand up to conventional surgery. As a result, there has been greater emphasis on applying minimally invasive techniques to this patient population. However, minimally invasive techniques are technically very difficult and have the disadvantage of poor/suboptimal visualization. To compensate for this limited visualization, there has been increased use of fluoroscopy and CT-guided navigation technologies. The latter allow correlation of patient intraoperative anatomy with preoperative anatomy in three dimensions, which may be especially beneficial in patients with metastatic disease, where tumor proliferation leads to disruption of normal anatomic landmarks. The objectives of this chapter are to provide an overview of the management of metastatic spine disease, to describe current intraoperative navigation techniques with a focus on intraoperative CT-guided navigation, and to describe how these navigation techniques can be applied to surgery for spinal metastases.

17.2 Overview of Management of Metastatic Spine Disease

Unlike primary spinal neoplasms, which are commonly characterized by local disease only, all spinal metastases are stage IV systemic disease, by definition. As a result, the goal of surgery in patients with metastatic spine disease is very different than that in patients with primary neoplasms: mainly, symptom palliation and improvement in patient quality of life, rather than oncologic cure. In some rare cases, where the patient is confirmed to have disease isolated to the spinal metastasis, it may be acceptable to perform en bloc tumor resection, as some literature has suggested that it provides clinically significant survival benefit for these patients. ¹⁶ ^, ²⁸ Such patients represent the vast minority of patients though—2 to 3% per some estimates—and as such will not be considered here. ²⁹

For the majority of patients with metastatic spine disease—those in whom disease is systemic and surgery is palliative—there are four considerations that need to be made in surgical planning, emphasized in the neurologic-oncologic-mechanical-systemic (NOMS) assessment paradigm. ³⁰ The first of these components is expected patient survival (systemic assessment). As surgery in this patient population is designed to improve patient quality of life, surgery is restricted to only those patients who are expected to survive long enough to rehabilitate from surgery and therefore realize the functional and symptomatic benefits conferred by surgery. Current consensus is that in order for a patient to be considered surgical, they must have a residual life expectancy of at least 3 months, though more conservative groups recommend that life expectancy exceed 6 months. ⁸ ^, ¹⁷ ^, ¹⁹ ^, ²¹ ^, ²² ^, ³¹ ^, ³² ^, ³³ ^, ³⁴ ^, ³⁵ ^, ³⁶ ^, ³⁷ As can be seen, the consideration of a patient for surgery is then highly dependent upon the ability of the surgeon and rest of the oncologic care team to accurately predict patient survival. Previous literature has indicated that patient prognosis is negatively associated with male sex, ³⁴ ^, ³⁸ ^, ³⁹ ^, ⁴⁰ increasing age, ⁵ ^, ¹⁰ ^, ²² ^, ³⁴ lung ⁵ ^, ³⁴ ^, ⁴¹ ^, ⁴² or GI primary pathology, ⁴⁰ polyvertebral disease, ⁵ ^, ¹⁸ ^, ⁴¹ ^, ⁴² ^, ⁴³ ^, ⁴⁴ visceral metastases, ⁵ ^, ²² ^, ³⁸ ^, ⁴¹ ^, ⁴² ^, ⁴⁵ ^, ⁴⁶ decreasing time between diagnosis of primary and metastasis, ⁴⁷ ^, ⁴⁸ and preoperative neurological deficit ⁸ ^, ³⁴ ^, ³⁵ ^, ³⁸ ^, ⁴⁴ ^, ⁴⁷ or nonambulatory status. ¹⁸ ^, ⁴³ ^, ⁴⁴ ^, ⁴⁸ ^, ⁴⁹ In accordance with these, several predictive scales have been generated to aid in patient selection, the most popular of which are the Tomita ¹⁶ and revised Tokuhashi scales. ⁵⁰ However, none of these scales has been demonstrated to predict patient survival across primary pathologies with great accuracy, ¹⁹ ^, ³¹ ^, ⁴⁶ ^, ⁵⁰ ^, ⁵¹ ^, ⁵² and so they are recommended as decision aids, as opposed to definitive treatment guides. ¹⁵ ^, ²¹ ^, ³⁷ ^, ⁵³ ^, ⁵⁴ ^, ⁵⁵ ^, ⁵⁶ ^, ⁵⁷ Additionally, some evidence also suggests that preoperative cachexia, assessed by body morphometry and fat distribution, may negatively predict postoperative survival (authors’ unpublished results).

The decision to operate on a patient with metastatic spine disease requires thoughtful consideration of the indications for surgery, potential benefits/risks of surgery, prognosis and likelihood of appreciable benefit from surgery, and patient expectations. ¹⁹ ^, ²⁰ ^, ²¹ Surgical indications may include the treatment of pain, spinal instability, and neurologic dysfunction. ³⁰ Metastatic epidural spinal cord compression with neurologic deficit, estimated to occur in 2.5 to 14% of patients with spinal metastases, ⁵ ^, ⁶ ^, ⁸ ^, ¹⁸ ^, ¹⁹ ^, ²⁰ ^, ³² ^, ⁴⁴ ^, ⁵⁸ ^, ⁵⁹ ^, ⁶⁰ ^, ⁶¹ ^, ⁶² ^, ⁶³ is perhaps the strongest indication for surgical decompression. These patients present with any of a variety of neurological symptoms, of which the most common are isolated or radicular pain (83–95%), ¹⁸ ^, ²⁷ ^, ³² ^, ⁶⁰ ^, ⁶³ ^, ⁶⁴ ^, ⁶⁵ sensory disturbances (50–70%), ¹⁸ ^, ⁴⁸ ^, ⁶⁶ ^, ⁶⁷ and weakness (35–75%), ⁹ ^, ¹⁸ ^, ²⁷ ^, ³¹ ^, ³² ^, ⁴⁸ though patients may also present with an inability to walk (11–68%), ⁸ ^, ¹⁸ ^, ²⁰ ^, ⁴³ ^, ⁴⁴ ^, ⁶⁰ ^, ⁶⁸ ^, ⁶⁹ autonomic dysfunction, and/or incontinence (50–60%). ¹⁸

From the late 1970s to early 2000s, the standard of care for these patients was radiotherapy, as class III evidence up to that point had identified it as being clinically equivalent to surgical decompression, but with lower associated morbidity. ⁸ ^, ¹⁸ ^, ³² ^, ⁷⁰ ^, ⁷¹ ^, ⁷² However, in 2005, Patchell and colleagues ⁸ presented level I evidence demonstrating superior neurological outcomes and superior survival in patients treated with surgical decompression and adjuvant radiotherapy compared to radiotherapy alone. As such, standard of care for patients with metastatic spine disease is now surgical decompression with stabilization where necessary, and adjuvant radiotherapy to address the tumor margins in radiosensitive pathologies. The extent of the margin that is left depends highly upon the radiosensitivity of the tumor (oncologic assessment). ³⁰ Radiosensitive tumors with high-grade epidural spinal cord compression, such as lymphoma, seminoma, and myeloma, can be treated with minimal decompression followed by stereotactic radiation (SRT) to the site of compression. The initial decompression prevents progression of neurological deficits and the radiation leads to rapid regression of the tumor. This more limited resection is known as separation surgery and is popular among frailer patients and those with relatively limited survival, as it has a lower associated morbidity. ⁷³ ^, ⁷⁴ ^, ⁷⁵ ^, ⁷⁶ ^, ⁷⁷ In contrast with the aforementioned pathologies, thyroid, colorectal, renal, NSCLC, sarcoma, hepatocellular, and melanoma metastases are considered radioresistant and require more extensive surgical decompression, as they are unlikely to regress significantly with SRT. ³⁰

Like neural element compression, mechanical instability in patients with metastatic spine disease can be treated with one of two interventions: surgical stabilization or vertebroplasty. ⁷⁸ Vertebroplasty may be advantageous for patients with borderline instability, limited life expectancy, or multiostotic disease not amenable to surgical intervention. However, in patients with gross instability who are considered healthy enough for surgery, surgical stabilization is the treatment of choice. Additionally, surgery is the only option for patients with involved levels lacking intact posterior cortices, as the intact cortical bone is necessary to prevent extravasation of the cement into the vertebral column. To help aid with this decision process, the Spinal Oncology Study Group developed the Spinal Instability Neoplastic Score (SINS), which categorizes the stability of the pathologic vertebra based upon spinal level (junctional, semirigid, rigid, or mobile spine level), pain (mechanical, nonmechanical, or no pain), lytic/blastic tumor quality, extent of vertebral body collapse, and extent of posterior element involvement. ⁷⁹ ^, ⁸⁰ Stabilization is recommended in lesions presenting with significant vertebral body collapse, mechanical pain, and tricolumnar involvement, whereas medical management with possible vertebroplasty is recommended for patients with more limited or asymptomatic disease.

17.2.1 Minimally Invasive Surgery for Spinal Metastases

As stated, two of the most important considerations for surgical intervention in patients with metastatic spine disease are the expectations that the patient (1) will survival long enough to appreciate symptomatic relief afforded by surgery, and (2) the patient is medically healthy enough to undergo surgery. Patients with systemic disease are a complex population with increased risk for postoperative complications (i.e., pulmonary and liver dysfunction). ⁸¹ ^, ⁸² Two widely accepted means for risk stratifying this population include the Charlson Comorbidity Index (CCI) ⁸³ ^, ⁸⁴ and Anesthesiologist Society of America Score, ⁸⁵ ^, ⁸⁶ both of which assign a cumulative comorbidity to patients based upon their preexisting medical conditions. Higher scores on both of these indices have been previously associated with postoperative complications in patients with spinal metastases. ⁸³ ^, ⁸⁴ ^, ⁸⁵ ^, ⁸⁶ However, not all inputs utilized in scoring for these systems have been demonstrated to impact postoperative outcome, and neither system is designed specifically for patients with metastatic spine disease or spinal oncology patients, in general. A score specifically designed for this population was recently published by de la Garza-Ramos and colleagues, ⁸⁷ ^, ⁸⁸ who developed a frailty score specifically designed to address surgical risk among patients operated for spinal metastases. They found that patients with greater degrees of frailty had higher complication rate and a 30-day mortality of 25%. This significant risk profile may tilt the risk-benefit profile of surgical intervention in such a manner as to preclude frailer patients from being offered surgical intervention. However, many of these patients would undoubtedly derive clinical benefit from surgical management. For these patients, the best surgical option may be use of minimally invasive techniques.

There are two widely accepted techniques for minimally invasive spine surgery (MISS) in the treatment of spinal metastases: video-assisted thoracic surgery (VATS); and mini-open/minimal access posterior decompression. The latter ⁸⁹ makes use of a familiar, posterior midline approach utilizing a 2-cm incision as opposed to a more conventional 5-cm incision. VATS is similarly related to a familiar approach—thoracotomy—albeit differing by the use of indirect versus direct visualization. As these techniques are still relatively new, the majority of reports have been published within the past decade. Evidence comparing minimally invasive and open approaches is expectedly limited, with only a handful of reports directly comparing open and MISS procedures. ²⁰ ^, ⁹⁰ ^, ⁹¹ ^, ⁹² ^, ⁹³ ^, ⁹⁴ ^, ⁹⁵ ^, ⁹⁶ ^, ⁹⁷ However, a recent review of this evidence by our group ⁹⁸ found the minimally invasive techniques to offer similar outcomes in terms of neurological recovery and pain improvement, while having shorter operative times, lower complication rates, shorter operating times, and shorter lengths of stay (Table 17‑1, Table 17‑2). Despite this apparently favorable complication profile, minimally invasive techniques are not without their drawbacks. First, minimally invasive techniques, especially VATS have steep learning curves, and second, the smaller incision and operating corridor employed by these techniques reduces visualization of the spine pathology. Intraoperative fluoroscopy can aid in indirect visualization by assisting with assessment of intraoperative instrument location. Moreover, it can be used to assist in vertebral instrumentation. But, given that many of the patients operated for spinal metastases have epidural cord compression as their primary indication, the key endpoint to surgery is adequate decompression of the neural elements. Such decompression generally requires expansive visualization, which can be challenging with minimally invasive approaches. Additionally, the soft tissue mass compressing the neural elements is not visible using standard fluoroscopy and so determination of adequate decompression in non-navigated MIS procedures is based upon surgeon experience. To overcome this issue and allow for superior outcomes, CT-guided navigation can be employed, effectively expanding the visible window in MIS procedures. CT-guided navigation may also aid in the surgical treatment of complex spinal pathologies, or in patients with anomalous anatomic structures.

**Table 17.1** Summary of studies directly comparing minimally invasive and open approaches for operative management of metastatic spine disease
Study	Technique	n	Operative			Clinical
Study	Technique	n	BL	OT	LOS	NI	PR	CR
Chou and Lu (2011) ⁹⁰	Open	5	3,120	408	–	100%	–	20%
Chou and Lu (2011) ⁹⁰	MIS	5	1,320	468	–	100%	–	20%
Fang et al (2012) ⁹¹	Open	17	1,721	403	–	76.5%	7.2	11.8%
Fang et al (2012) ⁹¹	MIS	24	1,058	175	–	91.7%	6.6	29.2%
Hansen-Algenstaedt et al (2017) ⁹²	Open	30	2,062	220	21.1	33.3%	5.6	40.0%
Hansen-Algenstaedt et al (2017) ⁹²	MIS	30	1,156	191	11.0	20%	5.2	23.3%
Hikata et al (2017) ⁹³	Open	25	714	189	–	56%	4.6	44%
Hikata et al (2017) ⁹³	MIS	25	340	205	–	56%	4.3	12%
Huang et al (2006) ⁹⁴	Open	17	1,162	180	–	70.8%	–	23.5%
Huang et al (2006) ⁹⁴	MIS	29	1,100	179	–	69.2%	–	20.7%
Kumar et al (2017) ⁹⁵	Open	18	961	269	13	50.0%	3.5	16%
Kumar et al (2017) ⁹⁵	MIS	27	184	253	9	56%	5.2	3%
Lau and Chou (2015) ⁹⁶	Open	28	1,697	414	11.4	42.9%	–	21.4%
Lau and Chou (2015) ⁹⁶	MIS	21	917	452	7.4	42.9%	–	9.5%
Miscusi et al (2015) ²⁰	Open	19	900	192	9.3	63%	–	0%
Miscusi et al (2015) ²⁰	MIS	23	240	132	7.2	65%	–	4.3%
Stoker et al (2013) ⁹⁷	Open	4	1,250	518	24	–	–	100%
Stoker et al (2013) ⁹⁷	MIS	4	813	367	5.8	–	–	100%
Abbreviations: BL, mean blood loss in mL; CR, complication rate; LOS, mean hospital length of stay in days; MIS, minimally invasive approach; NI, percentage of patients improving by 1 or more ASIA/Frankel grades following surgical intervention; OT, mean operative time in minutes; PR, mean pain relief in points on numeric pain rating scale. Source: Pennington et al (2018). ⁹⁸

**Table 17.2** Meta-analysis of studies directly comparing MIS and open approaches to management of metastatic spine disease
Endpoint	MIS		Open		Result	p
Endpoint	n	N	n	N	δx̅	p
BL (mL)	115	6	100	6	−608.3	<0.00001
OT (min)	115	6	100	6	−69.6	<0.00001
LOS (d)	61	3	52	3	−0.55	<0.0001
PR (NRS pt)	106	4	90	4	0.12	0.66
Endpoint	n	N	n	N	OR	p
NI	130	9	122	9	0.98	0.94
CR	188	9	164	9	0.58	0.05
Abbreviations: BL, mean blood loss in mL; CR, complication rate; δx̅, difference in means between open and MIS groups (δx̅ = MIS – open); LOS, mean hospital length of stay in days; MIS, minimally invasive approach; n, number of patients in group; N, number of studies in group; NI, percentage of patients improving by 1 or more ASIA/Frankel grades following surgical intervention; OR, odds ratio (>1 = more common in MIS group); OT, mean operative time in minutes; PR, mean pain relief in points on numeric pain rating scale.